Can cellular phototoxicity be accurately predicted on the basis of sensitizer photophysics?

The phototoxicity of three structurally related photosensitizers (PS), deuteroporphyrin IX (DP) and monobromo (Br-DP) and dibromo (Br2-DP) derivatives, was studied in murine L1210 leukemia cells. These compounds were chosen on the basis of heavy-atom-induced differences in triplet yield, phi T, and lifetime, tau T, and used as tools to test a model for phototoxicity based on photophysical parameters. All three porphyrins were found to localize preferentially in the plasma membrane of L1210 cells by confocal fluorescence microscopy. A poor correlation was observed between the measured photodynamic efficacies of these PS and a model using photophysical parameters determined by laser flash photolysis in homogeneous solution. However, an excellent correlation was obtained when the same parameters measured directly in the cells were used. The biological microenvironment of the porphyrins in cells induces significant changes in the photophysics of the PS. Reduction in fluorescence yield, phi F, and phi T observed for Br2-DP in cell suspensions arises from self association of the molecule due to increased hydrophobicity and high local concentrations. The photophysical model was also tested for its ability to handle variations in the oxygen dependence of cellular phototoxicity of these PS. The good correlation achieved between laser flash photolysis data determined in cells and the measured phototoxicity under air, 1.5% and 0.5% O2-saturated conditions, proves the intermediacy of singlet oxygen. This study gives further credence to the direct use of photophysical techniques to elucidate photochemical mechanisms in biological media while highlighting the potential pitfalls of using solution data to predict photosensitizing potential.